The long-term objective of the proposed research is to understand how the nervous system controls behavior. This question has significant implications for mental health, for mental illness is, ultimately, a disease of behavior. Taking a model systems approach, we will study the neuronal basis of behavior in the nematode worm Caenorhabditis elegans. The existence of a complete description of the synaptic connectivity of the 302 neurons of the C. elegans nervous system, and the wide range of genetic and physiological techniques for linking genes and neurons to behavior in this animal, makes it unusually well-suited to the proposed studies. In addition, more than 60% of all human disease genes have a correlate in C. elegans, including genes implicated in devastating neurodegenerative conditions such as Parkinson's, Alzheimer's, and Lou Gehrig's disease. The present proposal focuses on chemotaxis -locomotion oriented to a chemical gradient -a simple example of the important problem of how sensory information is transformed by movement control centers to generate adaptive behavior. We have previously shown that chemotaxis in C. elegans follows a well-defined rule: turn more frequently when going down the gradient, and turn less frequently when going up the gradient. Other researchers have identified a network of neurons that controls the decision to go straight or to turn, but it is not yet known how this network functions in chemotaxis. We will address this question using a remarkable combination of genetics, electrophysiology, calcium imaging, and behavioral analysis to test a mathematical model of locomotory regulation. The proposed research is significant because locomotion defects are one of the main consequences of altering the function of neurodegenerative disease genes in C. elegans.